Latest / Normal Men / Episode 56 - Bottom Feeders

Transcript

Episode 56 - Bottom Feeders · Normal Men

• 0:00
  Ed: Hey folks, Ed here. One quick message before the train wreck gets rolling. I am currently packing for a trip to Japan. I'm going to be gone for two weeks. The reason I'm saying this is yes, I'm bragging, but also this episode is a little lighter on the edit than usual. We are trying some new interesting stuff that Proctor has been working on to automate some of it. So we're interested in how that listening experience is for you. But the big warning is that this episode is unbleeped. Yes, children. That means you're going to hear British and other four letter words. If there are children or worse, the English listening, you may need to cover your ears, that's all. Enjoy the show.
• 0:55
  George: To a single ball. You are fucking up. Like you should be winning every run out and I'm beating you regularly. So what are you doing? But yeah, so that, that there's that.
• 1:14
  Lowtax: Yeah, that sounds tiring.
• 1:17
  George: Yeah.
• 1:20
  Lowtax: Rogue fish and chips shops,
• 1:26
  Propter: they're a real problem, I hear.
• 1:28
  Lowtax: Oh, they nicked me. Had it Ken it.
• 1:33
  Ed: Oi, Br.
• 1:35
  George: What you all about with the haddock? There's catfish. You can't be giving me catfish in my fish and chips.
• 1:42
  Lowtax: Yes, Headline, headline from the Times of London. Chippies caught selling catfish instead of cod and Haddock. Investigation found three in 10 businesses sampled were selling the cheaper species which is usually imported from Southeast Asia without telling customers.
• 2:02
  Propter: If only, if only we had a word that described. When you hold yourself out to be something that you are not for purposes of gain.
• 2:12
  Ed: Never heard scientists.
• 2:14
  Lowtax: Scientists use DNA testing to identify cases of catfish species being wrongly sold or described to customers in chippies.
• 2:25
  Propter: So like fish fraud is actually apparently a significant problem.
• 2:28
  George: Oh no, it really is.
• 2:29
  Propter: Yeah, that. But, but normally it's a, it's more of a high end problem that, that it's very difficult to confirm that the fish you're buying is actually the fish that is being sold and therefore restaurants sell fish that are not the fish that they're describing all the freaking time. And you know, whether or not this is a real problem depends on how much you care about whether or not the specific fish you're eating is the specific fish that you think you're eating. Which for some people is probably a lot and for other people is probably not that much.
• 3:01
  Lowtax: It is very important to me personally that when I am buying a food item that is wrapped in an old newspaper, I know that I am getting the, the correct kind of fish.
• 3:19
  Ed: You know what, minimally this is actually like not nothing in terms of like what you're eating, like, allergies. There are fish allergies that can be different from species to species. I don't know if catfish itself is one, but, like, there's some that my wife has, like, a bad reaction to.
• 3:40
  Lowtax: Oh, I agree that this is. This is not good. And the, the chippies should be punished for this.
• 3:49
  Ed: You know what should do?
• 3:50
  George: The Brits should join a multinational alignment of countries famous for very specific regulation around what goes into seminal foods. Right. Like that. That would solve this problem. And now I'm going to take a big swig of coffee and look at a newspaper from 2016.
• 4:11
  Ed: Yeah, but, like, here's, here's the really cool bit. Then they can leave.
• 4:17
  George: That's how you get catfish in your fish and chips.
• 4:19
  Propter: Yeah. So, I mean, this is a big deal.
• 4:21
  Lowtax: This is like, this is like 9, 11 for Tories.
• 4:27
  George: Dude, forget Tories. The reform is much more concerned about this than the Tories. There are not that many Tories.
• 4:33
  Ed: They don't know they can't read. I mean, come on. They make Mark Levin look like a vertebrate. That was a Prompter joke. Thank you.
• 4:46
  Propter: I appreciate that one.
• 4:47
  George: I feel like Quail is more the Tory thing.
• 4:51
  Ed: No, like, Quail, for his faults. He was at least like a person and not just like a skin draped over a robot made entirely out of racism. Like, there's a, like, you have to, like that era of Republican, like. Sure. Were they bad people? Yes. I, I.
• 5:09
  George: Sorry, I was.
• 5:11
  Propter: We're having two completely different conversations.
• 5:13
  George: I wasn't talking about Dan Quail. I was talking about the game bird.
• 5:16
  Ed: Oh, I thought you were still. I like my.
• 5:21
  Propter: Devour Dan Quail.
• 5:23
  George: Yes.
• 5:23
  Lowtax: If toys want to devour Dan Quail, that's fine.
• 5:26
  George: Too
• 5:28
  Ed: odd.
• 5:29
  George: But, yeah, go knock yourself out, fellas.
• 5:32
  Ed: I'm. I'm actually getting a little bit of legs.
• 5:36
  Propter: Quail.
• 5:37
  Lowtax: Quail is, is more of a Tory thing, but it depends on the type of Tory.
• 5:40
  George: Right.
• 5:41
  Lowtax: Like, are we talking Tories who eat at the Fat Duck, or are we talking about rural white trash Alabama or England? Tories. Right. The ones who think that, you know, kebabs are going to turn their kids into transgendered Muslims or something.
• 6:00
  Propter: Again, I feel, I feel obligated at this point to mention that fish and chips are one of kind of like the paradigmatic examples of immigrant cuisine that this is that when fish and chips, which we now think of as being the quintessential British fast food available.
• 6:18
  Lowtax: It's Jewish.
• 6:19
  Propter: Yeah, yeah. It's Sephardic Jews from Spain and Portugal. And for years, it was referred to as fish served in the Jewish style.
• 6:28
  George: Yep.
• 6:29
  Propter: My other conclusion here is that the Sephardim have better food than the Ashkenazim.
• 6:33
  George: Oh, man. Oh. I don't know if we can just drop that kind of nuclear weapons grade take into the podcast. You may be onto something. But I will not comment on that though, because I don't need to start over.
• 6:50
  Lowtax: I mean, I think the, like, we're talking about Central Europe versus the Mediterranean. So that's a terribly. That's not a terribly hot take, frankly.
• 7:00
  Propter: Yeah.
• 7:02
  George: But so objectively you're right. But the whole separate IM Ashkenazi thing, there are cultural valences at work here that we need to be careful about getting involved in that particular fight.
• 7:17
  Propter: Gentlemen, we are not the podcast to go deep on this. Maybe, maybe we'll get a guest on who can talk about this more intelligently.
• 7:24
  Lowtax: Are you drinking Nuka Cola?
• 7:26
  George: Yeah. Sipping on a Nuka Cola, which for those aren't aware, there was a really good tie up between Joan Sotako, the Northwest darling soda artisanal soda producer, and the Fallout video game series, I think exclusively available at Costco, if I remember correctly.
• 7:42
  Ed: Are they still artisanal when they're being sold at Costco?
• 7:46
  George: I'm just loosely using the term artisanal,
• 7:48
  Propter: but sure, the artisans are locked in the factory, but yes, still artisanal.
• 7:54
  George: Also, Nuka Cola is bright blue. Like, not North Carolina, like unc, light
• 8:00
  Lowtax: blue, like neon blue.
• 8:01
  Ed: It is blue raspberry blue.
• 8:04
  George: Yes, yes, yes, yes, yes, yes.
• 8:05
  Ed: Yeah. Which is neither raspberry nor any other kind of berry.
• 8:09
  Propter: Correct.
• 8:10
  George: For my money, the Nuka Cola was the second best in the pack. There was. There was a grape one that was just. I mean, it was grape soda.
• 8:16
  Propter: Which.
• 8:17
  George: Grape soda isn't the worst tasting stuff you can put in your mouth, but it's. It's not ever going to be particularly good. Cola was like not. But the Sunset Sarsaparilla and that.
• 8:30
  Propter: Yeah, yeah, that was.
• 8:32
  Ed: I like.
• 8:32
  Lowtax: Yeah, I liked both the sarsaparilla and the Nuka Cola a lot.
• 8:35
  Propter: The.
• 8:37
  Lowtax: The grape was like you said, it's whatever. It's grape soda. Then there was. What was it like apple or something like that?
• 8:44
  George: I only had three in my pack.
• 8:46
  Lowtax: Yeah, I had. I think I had four.
• 8:50
  Propter: I don't remember.
• 8:51
  Lowtax: I think there was an apple one and it sucked.
• 8:54
  George: Joan Sodako does a green apple soda.
• 8:57
  Propter: Yeah.
• 8:57
  George: Or at least it did. It's pretty good back in the day,
• 8:59
  Propter: but yeah, we stay away too.
• 9:00
  Ed: Yeah.
• 9:01
  Propter: Yeah.
• 9:01
  George: They also do a very on brand for Proctor to be drinking that particular geographic valence of soda. Good job, buddy.
• 9:07
  Propter: Yeah. Jones. Jones also does an annual Thanksgiving pack, if you're into that, which is a pack of sodas that taste like Thanksgiving foods.
• 9:17
  George: Savory or sweet?
• 9:18
  Propter: Thanksgiving. Yes, Savory. Yeah, yeah, yeah, yeah, yeah. There's. There's a boy turkey soda, gravy soda, you know, Mac and cheese soda.
• 9:26
  George: No. Have you had it?
• 9:29
  Propter: I tried. I think it was the gravy one once. Once.
• 9:35
  George: And. Yeah.
• 9:38
  Ed: Try it a second time.
• 9:40
  Propter: Yeah, it tastes kind of like gravy, but it's soda. If that's something that appeals to you, then it's probably the kind of thing you'd like.
• 9:48
  Ed: You just carbonate your gravy. It's fine.
• 9:51
  Propter: Yeah, like, I mean, like, it's. It's obviously a stuffed food, right? Like. Like. Like. The point of this is not to enjoy it. The point of this is to. Is to drink the soda that tastes like Thanksgiving.
• 10:01
  Ed: How to carbonate gravy soda?
• 10:04
  George: Steam and go to homemade gravy, obviously.
• 10:06
  Lowtax: Do the proceeds go to charity or something? Is that the. Is that the gimmick?
• 10:12
  Propter: No, I don't think so. I think they just. I think. I think they're selling, so this is just.
• 10:15
  Lowtax: They're just assaulting people for no particular reason.
• 10:19
  Propter: So it's got, like. It's got. I guess it's like a kind of a collector's thing. Like, it's a limited edition every year, and they change it up every year. So.
• 10:29
  Lowtax: Where is Jones Soda company located?
• 10:31
  Propter: Washington state, baby. Okay.
• 10:35
  Ed: Actually, I think you couldn't really carbonate gravy because it'll get, like, warm liquids off gas faster. So, like, you're gonna have to do gravy. And that's what you need to do,
• 10:46
  George: is you need to. You take the gravy and you dilute it a bit, and then you let it sit there for a while, and then you run it through a centrifuge, and then you take what's left. The gravy part of the top.
• 10:57
  Ed: Yes.
• 10:57
  George: Y. So, like, all the solids come to the bottom and what's left? I had a friend, now, dearly departed friend, who was a bit of a nerd about cocktails and related subjects, and he actually bought a centrifuge to do stuff like this because, as one does. And, yeah, you can. You. You can make clarified taste liquid of just about anything if you try hard enough, which apparently the people Jones Soda do, so.
• 11:28
  Ed: So wait, do you guys not have centrifuges?
• 11:31
  George: No.
• 11:32
  Ed: Yeah, it cost about $80. AliExpress is an amazing place.
• 11:37
  George: Why do you have it? Don't answer that. We don't want to get subpoenaed.
• 11:40
  Ed: It's. It's for woodworking.
• 11:43
  George: Why do you need a centrifuge?
• 11:44
  Ed: No, it actually is like because you can separate out like you can separate out dice and, and like, and finishes and people do really stupid things. I'm not saying that this is a useful thing for woodworking. I'm saying that's why I have one.
• 11:59
  George: But that was implied for what?
• 12:01
  Ed: You get to see it go. It's. It's a good time.
• 12:05
  George: Oh man.
• 12:07
  Ed: They are very cheap, as are most other bootleg scientific pieces. Scientific equipment is, is remarkably affordable if you are willing to take certain safety anti precautions.
• 12:20
  George: I don't know if I want something that's doing a thousand plus RPM for 80 bucks. Like that feels spiritually.
• 12:26
  Ed: Can I introduce you to a power drill?
• 12:29
  George: Yeah, but that's not got. I mean that, that there's the radius you're working with there and therefore the velocity of the outer edge of whatever you're dealing with is.
• 12:42
  Ed: Can I introduce you to a hole saw?
• 12:44
  Propter: Yep.
• 12:47
  George: It's not made of glass though. Aren't you doing the centrifuge?
• 12:50
  Ed: That's true. It does use glass tubes. But like they're. You have to get ones that fit specifically into like the molded slots so they can't move around or anything.
• 12:58
  Propter: Okay.
• 12:59
  Ed: It's. It's not. It's not in the top five of most dangerous things in my garage.
• 13:04
  Lowtax: Proctor, you're muted.
• 13:06
  George: Yeah, very muted.
• 13:09
  Propter: On the one hand, Ed is making the case that this is nowhere close to the most dangerous thing he has in his house. And on the other hand.
• 13:18
  Ed: No, I'm just talking about the garage.
• 13:20
  Propter: And on the other hand, Ed's, Ed's. Ed's garage. I assume his whole house is, is a, is a rococo house of horrors that like we get. We get the hazmat reports from Ed occasionally on his. On his off gas risk which he has minimized. To be fair.
• 13:36
  Ed: I, I don't have any off gas risk aside from wood finishing. Right. I don't print an abs. I, I keep all of that stuff out of the house.
• 13:45
  Propter: I.
• 13:45
  Ed: When I get the, the external garage done, that's going to change some things because a, A load bearing part of the external garage is a rated fume evac situation.
• 13:57
  George: I hope you're not literally going to use that as a load bearing part
• 14:00
  Propter: of the external garage.
• 14:02
  Ed: Yes, we're going to just balance it, like the. The center roof post goes straight through the vents. It. It's going to work great.
• 14:14
  George: All right, well, how did we get to this from.
• 14:17
  Lowtax: How did we get to this from fish and chips?
• 14:20
  Ed: Because I was trying to carbonate gravy right.
• 14:24
  Lowtax: As one does.
• 14:25
  Propter: Gosh, yeah.
• 14:28
  George: Okay, enough clowning around. We're done with all that. This is Normal Men, a podcast from four men who are clinging desperately to normal normalcy in an era when that's asking a bit too much.
• 14:42
  Propter: I'm George In Charlotte, in D.C. i'm
• 14:45
  Lowtax: Proctor Malone in Outer Florida. I'm low tech Speedrun enjoyer.
• 14:49
  Ed: And I'm Ed in Boston.
• 14:56
  George: What happened this week, fellas? Anything interesting?
• 14:58
  Ed: No. That's the end of the episode. Bye, everyone.
• 15:02
  Lowtax: Jay Powell is never leaving the Federal Reserve.
• 15:06
  George: Yeah, just. If you've seen the Wolf of Wall Street, I ain't leaving. Yeah. So just to recap where we're at, the Fed chair's term expired. It just expired. And so the Trump administration gets to appoint a new Fed Chair. They need someone either on the Board of Governors currently or they need to replace Governor Miron with a. Whose term technically expired in January and can now be replaced by Congress. Replace him with the new chair. If Powell were to follow the typical, not the exclusive, but the typical path of former Fed chairs, he would step down from the Board of Governors and resign his seat. He's under no obligation to do so until 2028, when his term ends this week at the FOMC after about a year. I would say of varying degrees of speculation around what he would do. He said, well, I'm going to welcome Chair Warshaw onto the committee, and I will not be leaving for the foreseeable future. What does that mean? Could be a few months, could be yours. We don't know. My money's on yours, but it's contingent. It's entirely possible he steps down earlier, but I suspect that the reason he's not stepping down, specifically the legal attacks on the Federal Reserve, will not end anytime soon. And in fact, we also got news this week that Judge Jeanine, the U.S. attorney for the State or for the District of Columbia, is continuing to appeal a District Court order that was barring subpoenas delivered to the Federal Reserve up to, I think. I think en banc, D.C. circuit is next. I can't remember the exact procedural posture, but she's continuing to appeal those subpoenas being quashed because, quote, unquote, the principle of the thing, Basically the idea that District courts shouldn't be able to Decide whether a subpoena is legal. Just nuts. But yeah. So the Justice Department said, we're done with the investigation into Chair Powell. Specifically we're going to pass it over to the Fed oig, which by the way, the Fed OIG investigating cost overruns on a Fed building project is completely reasonable. And the fed OIG is like 15 plus year public servant who was the DOJ OIG for over 10 years, appointed it under Obama and stuck through the whole first Trump administration and through Biden and now it's the Fed oig. So very unlikely it's going to be anything other than a fair investigation there. But the DOJ is still putzing around while specifically Judge Jeanine is putzing around. And so Powell said, no, I'm going to stay on. And the irony of this is that because Mirren therefore has to leave to make room for war, the FOMC actually shifts more hawkish with the Fed's Fed chair switching from Powell to Warsh, which is very funny.
• 18:13
  Propter: Mirren is, is, was originally a Trump appointee.
• 18:17
  George: Right? Or Trump appointee. Yeah, he. So he was appointed to fill a seat that opened up by Coogler. Coogler was set to leave the fomc, or Board of Governors and FOMC in January, but she resigned last. It was either August or September. Later reporting suggested that she left because her husband had been engaged in a series of, let's just put it this way, transactions he shouldn't have been engaging in. I don't think there's any evidence to suspect that Kukler herself did anything wrong, but there's a level of impropriety around a husband day trading, essentially. And ironically the husband is a MAGA guy. So.
• 19:04
  Propter: Yeah, there's also, there's also memory serves, there's. There wasn't any evidence that the husband was, was like particularly making money off this. He just now he was just pumping stocks. I mean, like, he shouldn't have been doing this. There was no question he should not have been doing this given his, given his wife's position. But this is a story that was maybe less than it should have been. On the other hand, this is a story that in retrospect probably gave the Trump administration the idea to start going after dirt on all the people on the Fed board that they wanted to get rid of.
• 19:32
  George: Anyways, I don't think the underlying details mattered, but because I think Cook was first Anyway. So this was, this was. She left after the attacks on Cook began.
• 19:43
  Propter: Yeah, I mean, we don't really know how this is percolating in the background though, because at the time that we got the Coogler resignation, the story wasn't public at all. It wasn't clear why she, why she had stepped down, when she stepped down.
• 19:54
  George: I'm just saying, like the attacks on Cook started. I mean, so the attacks on Cook were dragged out by fha. Fha, by Pulte. He, by the way, like we've talked about this before, those in that investigation by Pulte where he pulled mortgage records related to Cook, almost certainly broke laws. We don't have any evidence that anyone else did something similar to pull. Like for instance, the transaction records that would have shown the day trading by the husband. Right. That it strikes me as most likely that he went after the black woman first and the Latina woman second. But you know what? Six of one, half dozen, the other anyhow there. So, yeah, Warsh coming onto the committee. Trump thinks the Fed chair sets interest rates. He is very mistaken in that particular point of view. And it was actually funny at the FOMC last week, we had three dissents. Basically three people wanted to say, look, we could either hike or cut next. It depends what happens. And instead the statement for the meeting did not say that. The statement from the meeting said, implied that they're still sort of easing bias. We'll cut when we can next, but we're not going to do that near term. Okay? No mention of hikes. Whereas the three people who said we don't agree with this, they said we're fine with holding rates at this meeting, which is the outcome from this meeting. But we want language that suggests that we could either hike or cut next meeting or in the future. And there is a very clear signal there from everybody on the FOMC who is not a Trump appointee or Powell, that they're not just going to do what Porsche wants.
• 21:33
  Propter: So, yeah, this is, this is not a common situation.
• 21:38
  Ed: Right.
• 21:38
  Propter: I mean, typically the chair is understood to be speaking for the Fed, even if they're not making the decisions. Their job when they, when they speak to the media is to convey the sentiments of the Fed. And it seems increasingly unlikely that that's going to be what Warsh is doing.
• 21:52
  George: Well, okay, so formally, the FOMC is not ever just the chair deciding stuff. In the 1990s, it sort of worked that way with Greenspan because he had a lot of political power within the fomc. But that was not, that was not a structural thing. That was not like how the FOMC was organized. They just acted that way under Bernanke Yellen, and to the greatest extent, Powell. Powell has been really, really clear every time. Like, this is a committee where, you know, certain people on the committee feel such and such way. Certain people on the committee feel another way. Yellen said this all the time as well, and Bernanke said it as well. But really, Powell has been clear about this, that he is the guy sitting in the middle kind of trying to get everyone on the same page to come to a consensus. And Warsh may try and take a similar approach, but if he does, we're not going to be cutting rates anytime soon. And if he doesn't, then it's very possible that the Fed chair dissents from the committee action.
• 22:59
  Propter: Right. I mean, it's going to be really funny. Put it this way. Trump, Trump didn't appoint Warsh because he thinks that Warsh is going to do the same thing that Powell has been doing in terms of, in terms of sitting in the middle there in terms of trying to evoke consensus.
• 23:12
  George: Right. But that's also because Trump is an idiot and thinks that Powell sets rates and Warsh will set rates, and that neither of those things is true in this.
• 23:21
  Propter: Right.
• 23:21
  Lowtax: Which of course means that he's going to absolutely torch Warship down the road.
• 23:26
  Ed: Yeah, like five minutes.
• 23:28
  George: Oh, God, it. He's going to, like, send a Secret Service to Worship's house.
• 23:34
  Ed: Yeah.
• 23:34
  George: So Wash made a deal with the devil. And oh, there's one other note here. So Governor Waller, who is a Trump appointee, but yeah, he's pretty reasonable guy. Like, I disagree with his perspective sometimes, but whatever he skewed really, really dovish until Wash was appointed. And then his first speech after Warsh's formal appointment process got underway was ahead of this last FOMC meeting before the blackout period. The Fed doesn't talk to the media for two weeks before the meeting or a week and a half before the meeting just to sort of keep clear for them ahead of the meeting. Anyhow, right before that, like the last Fed speech before the blackout period, was Waller being like, oh, yeah, actually I'm not dovish anymore and I'm really worried about a series of further inflation shocks and the labor market actually looks fine. And forget everything I said for the past six months.
• 24:22
  Lowtax: So we love a messy bitch.
• 24:27
  George: So it's like, it's not just, oh, you're losing the most dovish guy on the committee. Mirren, who has basically been saying, I don't care what core inflation's doing, I don't care what the unemployment rate's doing. We need to cut 600 basis points like you're losing him off the committee. Warsh will be dovish, but not that dovish. Waller has shifted and Powell is now being very clear that he's like, I'm not going to be as restrained. I'm going to vote kind of different from that now. So, yeah, just spectacular stuff from the President, folks. They're calling him the greatest brain of all time.
• 25:02
  Propter: Yeah, I mean, I think that stories like this where you see these institutions that historically have done their deliberation relatively behind closed doors and tried to put on at least the impression of consensus in their public proclamations, or if not consensus of very delineated disagreement, like the Fed, like the Supreme Court, where we're now getting maybe more of a sense of some of the occasionally interpersonal dynamics that happen behind closed doors. And it seems likely we're going to see more of that from the Fed than we have in past years with this particular upcoming Fed.
• 25:41
  George: So I think an important thing to keep in mind is that economics is not anywhere close to, as an adversarial thing as the law is. So when you get a group of economists or just people that look at the economy all day on in a room together, right. And like Powell is a really good example of this. Right. He's not a trained economist. He's a private sector guy who ended
• 26:09
  Lowtax: up, he's a lawyer, I think, isn't he?
• 26:11
  George: Yeah, yeah. He came up through private equity. And it's, you know, it's a long story. And there are a lot of other regional Fed presidents that are kind of in that similar boat who are not Ph.D. economists. Right. But when you get people who look at financial markets, the economy all day and especially have to forecast them, you tend to get a. When they're talking about the economy, these are people who are not going to go to the absolute mat over a binary outcome because they know they could be wrong and frequently are. And so they have their forecast, they have their views, they have their disagreements. You see dissents on the FOMC all the time. Three dissents is unusual meeting this past week, but by all means disagreement is normal and there are back and forths in public speeches. But it is not like the Supreme Court because they all in general understand they're rowing towards the same goal and the adversarial ness is just lower. There are some exceptions to that, but in general it's really easy to see these people as colleagues as opposed to combatants. And the hawk dove battles, rhetorical battles that get framed are very Often a media creation, especially as an oppositional thing. Whereas on the Supreme Court. No, like they are actually debating and trying to win an argument. Right? They are, they know there's only one winner and one loser for each case. And so that's how they're approaching everything.
• 27:43
  Ed: So imagine, imagine trying to debate something with Clarence Thomas.
• 27:50
  George: Really rather not. Yeah. So I think regardless of what energy war springs to the picture, I will also say even Mirren, who is just so far out of consensus, like completely like bonkers takes. Right. I've heard from people inside the Fed who are like actually like he's very collegial. Like he wants to have access to the right resources, he wants to hear, hear people's opinion. Like his ideas are kind of nuts, but like he's not some lunatic about it. Right. Which is a really different thing. Like Mirin is the Fed equivalent of Alito or Thomas. Right?
• 28:33
  Ed: Oh boy.
• 28:34
  George: Yeah. No, I mean it's, it's not in terms of like, vibe, but just in terms of like how far to the right his views are in this, in this instance is to the right means close to Donald Trump. But yeah, so like it's really different, it's really, really different from the Supreme Court. And I actually suspect that the collegiality and the sort of like rump of the committee being on the same page and like having each other's backs is going to intensify and it's going to be like Warsh and you know, maybe Bauman kind of off trying to say whatever thing on the side and it's just going to be irrelevant and the rest of the committee is going to go, we're over here.
• 29:17
  Lowtax: So yes, Warsh will be fully integrated into the global Irish banking conspiracy.
• 29:26
  George: Yeah, I can't wait for Powell to run for president. 2028, man, we're just going to take that card out of the Canadians deck and run with it.
• 29:33
  Lowtax: No, we're not supposed to be appointed back to fed chairman in 28 or whenever Warsh's term ends.
• 29:42
  George: Well, no. So Powell's term on the Board of Governors ends in, I believe It's April of 2028. It's first half of 2028. So basically he'll have a, a time there will be another governor appointed. Well, I mean they might not get it done. It depends how Senate turns out in 2026, et cetera, et cetera. But like they're there. He, his term on the Board of Governors ends before Trump leaves the White House.
• 30:08
  Lowtax: Right. But could, could a future Democratic president appoint him again.
• 30:12
  George: Sure.
• 30:13
  Propter: They won't.
• 30:14
  George: Yeah, they won't. And the reason is Powell at this point. There was a good story by Nick Timoros in the Wall Street Journal today talking about how, like, he wanted to retire. He was like, I've done two terms as Fed Chair. Like that is. That is a. You know, the whole presidency ages you thing. Like, same thing with the Fed chair. He's got just as much responsibility. I mean, it's not as life and death day to day, like you don't have a nuclear football. But in economic terms, it's pretty close. And he's lived through some wild economic times. Right. So Powell doesn't want to be there any longer than he has to be. And I think he's probably, if you asked, if you like, in his heart of hearts, he's like, this is, man. Like, I just want to go drink a daiquiri. And this is making me stay on this fucking committee that I've been on for 15 fucking years.
• 31:05
  Ed: Yeah, that sounds about right. So he could fuck up the strain of Hormuz.
• 31:12
  Propter: Yep.
• 31:13
  George: Yeah, yeah. Speaking of which, no movement on that. That's our update on Iran still.
• 31:20
  Ed: They're still there. What do you want to bet that three weeks from now when I'm back, we will say the same thing?
• 31:27
  Propter: Yeah, yeah, yeah, yeah.
• 31:29
  George: Almost three weeks from now. If it's still closed three weeks from now. So just, I ran the numbers and it's like we're basically. We're missing 900 million barrels. You know, like, it works out to about 14 million barrels a day in global liquids. Global liquids demand is like 105 million barrels a day. So if you. And you kind of got to net some stuff out, and you've got Yambu and you've got the Emirati pipeline to the east coast, east side and whatever. And, you know, back of the napkin, it's roughly. We're missing 900 million barrels. Well, OECD commercial inventories coming in were like 2.8 billion barrels. So we're like a third of the way through that. And you can't run that to zero before bad stuff happens. Like you, like just operating, just sort of flowing stuff around. You need some pretty big cushion there. In other words, like, the straight has been closed for two months. It's not. We're not going to have $120 barrel Brent in two months. Right. If it stays close another two months, it's going to be. We're going to hit convexity at some point. Just rep. And I don't. I Don't think it'll be this week. I don't think it'll be next week, but maybe the week after that. Like it' point here we're going to hit the inflection point and it's no longer like, oh, is the straight open? Is the straight close? It's like, no, we don't have any barrels and we will pay 200 for them.
• 32:56
  Ed: We'll see.
• 32:57
  Propter: Anecdotally, it kind of seems like we're starting to see some movement in gas prices domestically. So that started to bite in about the past week.
• 33:07
  George: Yeah. Couple special factors exacerbating that. So the, the biggest thing is we've got some refinery outages in the Midwest and the Midwest is totally insulated from global oil markets. You know, the prices do get arbitraged, but they, they basically have captive Canadian crude that's flowing to them that doesn't really have anywhere else to go. And so Midwest refiners are running that plus some domestic US supply which go to the rest of the world. So there. But like the flip side of that is it's really hard to get supplies into the region. So when you have a couple of refineries go out, which is what's happened over the past couple of weeks, you see like really, really big inflection points in, in gas and diesel there. And that's what's happened. So there is special circumstances there. But yeah, it's only. It. Yeah. Crack spreads are still wide. They had, they had narrowed pretty. Yeah, I know. We did a whole episode on it. You can go back.
• 34:02
  Ed: It's.
• 34:02
  Propter: It's.
• 34:03
  Ed: For some reason I find global liquids almost as off putting, but for a completely different reason.
• 34:11
  George: So when we think of a barrel of oil, we think of like, I did it. Well, no, but I'm doing exposition here. It's unsettling. It looks unsettling too. I mean, crude oil is nasty stuff. But you think of a barrel of crude oil being like this uniform brown viscous liquid, right? And it's really like there's that, but there's also like liquefied petroleum gases which are, I mean, think something similar to propane. It's not propane exactly, but vaguely in the neighborhood. You also have this like really wide range of crudes where it's like some stuff is basically tar and other stuff is pretty close to refined fuels. Like it's, this is a big thing. So when we say like liquid petroleum supply, it that's like mostly crude oil and then some other stuff thrown on top. Natural gas liquids, liquid Flowing gases like that, that sort of stuff.
• 35:13
  Ed: I'm sure that's all like very important. I'm just over here thinking I really wish my batteries didn't want a thermal event right now.
• 35:20
  George: I mean you can get 80%. You're fine.
• 35:22
  Ed: Well, so funny enough, I got a letter yesterday saying that there is a dealer fix to modify the firmware or something. So I asked my dealer, does this mean this fixes the recall? They said no. So like that's how we're, we're doing this right now. 80% is like fine. Right?
• 35:43
  George: Have you talked to an attorney about it? I know that was on the menu.
• 35:46
  Ed: I did. And not great results. So I still have the car.
• 35:54
  George: Congrats.
• 35:55
  Ed: Yeah. Well, go Big Blue. Can we just like buy a BYD in this frigging country?
• 36:03
  George: I mean you wouldn't have had this problem if you'd just gone full electric to begin with and gotten a F150 Mustang or something.
• 36:09
  Ed: Yeah. Do you know what I would have had instead? Double the friggin car payment. You have no idea how cheap this car was.
• 36:19
  Propter: Sleep's coming.
• 36:20
  George: Now we know why.
• 36:23
  Propter: Yeah.
• 36:24
  Lowtax: Although watch, watch a Big Blue and others find out they have a battery issue on the fully electric ones too.
• 36:33
  George: I mean I'm 42000 miles in and I know there are people on the subreddit who. There was one guy I saw who had driven his Lightning like a hundred and fifty 000 miles in under two years. It was crazy. It was like unbelievable. People were doing the math and we're like how, how are you doing this? And no, no issues whatsoever.
• 36:54
  Ed: I don't.
• 36:55
  George: I mean that's great. Which is how electric cars should work on paper. But.
• 37:00
  Ed: Yeah, right.
• 37:01
  Propter: I mean one of, one of the big promises of the great electrification was then long term maintenance would be better because there's fewer moving parts. So we're not really my experience.
• 37:13
  George: I mean tires. That's it.
• 37:16
  Lowtax: I mean, I think so. I've seen rankings on vehicle reliability and the Lightning generally does not do very well. But at the same time like cars are really well made nowadays, you know.
• 37:32
  Ed: Oh yeah.
• 37:33
  Lowtax: Even really unreliable brands like with the possible exception of Land Rover and Tesla, you know most cars are probably going to get you 120 to 150000 miles at a bare minimum.
• 37:48
  Ed: Oh yeah.
• 37:48
  Propter: Like reliability is a complicated concept.
• 37:50
  George: Right.
• 37:51
  Propter: Like you've got, you've got sort of like.
• 37:52
  Lowtax: Right, right.
• 37:53
  Propter: Reliability or relatively low mileage and then you have how, how long can you drive this thing before Everything starts to fail and, you know, it's. It's a good thing if you have. If you have a car that you can drive a long way before everything starts to fail and then everything starts to fail at once. That's fine. That's the kind of intended end of life cycle for a car. Ideally, those things would be easy to replace, but. But, like, that's not necessarily a failure of engineering if you have things that. That start to fall apart eventually.
• 38:24
  Lowtax: Oh, do we have. We have breaking news from Ed.
• 38:28
  Ed: Oh, no, don't put this on me. I just happen to be. To be looking.
• 38:33
  Lowtax: You put it in the chat.
• 38:35
  Ed: Yeah. Well, thoughts and prayers, I guess, to Rudy Giuliani, who is currently hospitalized in critical but stable condition.
• 38:45
  George: We're thinking and praying about the stable part.
• 38:50
  Propter: We're thinking and we're praying. Let's not. Let's not get bogged down in details.
• 38:58
  Ed: So that happened.
• 38:59
  George: All I can imagine is that. Is that, like, Sting video of him, like, on his back with his feet off the bed, like, undoing his belt?
• 39:11
  Propter: That was. That was from Borat, right?
• 39:14
  George: Yeah, yeah, it was.
• 39:15
  Ed: Yeah.
• 39:15
  George: That was a Sasha Baron Cohen joint.
• 39:17
  Lowtax: Yeah.
• 39:17
  George: Oh, yeah.
• 39:18
  Lowtax: I can only ever think of him now with the hair dye dripping down his face when he was sweating outside of that lawn care place.
• 39:29
  George: Four Seasons Landscaping.
• 39:31
  Lowtax: Four Seasons Total landscaping. Yes.
• 39:33
  Ed: Yes.
• 39:35
  Propter: Some quality advance work by somebody there.
• 39:37
  Lowtax: He's an amazing guy. You know, like, 911 happens and he becomes, you know, America's mayor and all this bullshit, and people love him. He has huge approvals and all this stuff, and then he just becomes this weird little goblin creature in. At some point in, in the Trump era. And everybody, Everybody despises him now. It's amazing.
• 40:06
  Ed: I mean, they despised him on 9 10, too.
• 40:09
  Lowtax: Yes, that's true. But, I mean, all. All New York mayors are typically despised.
• 40:14
  George: Did he have really bad approvals before 9 11? I can't remember. I.
• 40:18
  Lowtax: Yes. I mean, remember, the New York mayoralty is the French presidency of the United States, right? You know, you get elected, no one's terribly happy about it, and then within five minutes, your approval rating is, like, negative. And everybody, everybody wants to throw you off a bridge and stuff.
• 40:42
  Ed: Mom, Donnie has lasted, like, eight minutes. So, you know, he's doing pretty well.
• 40:47
  Lowtax: He's. He's talented. Like, he. He might last, I don't know, three months, four months.
• 40:58
  George: I mean, 48% approval rating is not terrible. That's the most recent figure I could find from a quick Google. And that's not bad. That's Fine.
• 41:06
  Propter: Like, yeah, New Yorkers don't like anyone so famously.
• 41:12
  Lowtax: No.
• 41:12
  Ed: Yeah.
• 41:14
  George: Each other, any other city dweller, rural people. Forget it.
• 41:19
  Ed: The concept of politics entirely. Thank you, Chuck Schumer.
• 41:24
  Lowtax: Yes.
• 41:27
  George: Oh, boy.
• 41:28
  Lowtax: Although, George, you're, you're making a bold assumption there. Thinking New Yorkers would see anywhere else as a city.
• 41:35
  George: Yeah, that's part of it.
• 41:40
  Propter: So we've got some exciting cruise ship news I want to talk about briefly. This is either going to sound really stupid and in a couple days, by the time this episode comes out, or it's going to sound forward looking. There was a cruise ship going from Argentina to South Africa that had six people on it get sick with hantavirus, of whom three died.
• 42:01
  George: Tell us about the hantavirus.
• 42:02
  Propter: So hantavirus is normally a disease that you get from rodent droppings. It's endemic to a few places in the world. One of them is the Andes. And most importantly for our purposes, it is pretty lethal. You've got about a 2 in 3 chance of surviving a hantavirus infection. And it is not normally spread human to human. Notably, it is not normally a major problem shipboard. I haven't found anything that contradicts that yet. There are definitely rats who live on ships. But I haven't seen anything that suggests that hantavirus is like a major scourge of the seas or anything. That this is something where there's a real risk of you getting hantavirus at sea.
• 42:49
  George: Well, I mean, also because, like, there just aren't that many people on ships compared to how many there used to be. Right. Like if you used to, like the number of sailors relative to the amount of stuff we send around the world on ships is very, very low compared to history. Right. And the cruise ships are. Well, yeah, that's sort of a different issue.
• 43:09
  Propter: But I mean, if we don't think that there's human to human transmission, then that doesn't matter now. So it turns out that there is at least some evidence that the specific strain of hantavirus that lives in the Andes mountain has had some past human to human transmission. And again, this ship left from Argentina,
• 43:30
  George: so famously not in the Andes, famously
• 43:33
  Propter: not in the Andes, but the kind of place that if you were, for instance, a vacationing set of people who were coming from Europe to South America, you might do an Andean swing before you caught your cruise ship in Argentina.
• 43:48
  George: Sorry, I just realized like this, this is a cruise from Argentina to South Africa.
• 43:54
  Propter: From Argentina to South Africa, yeah.
• 43:56
  George: This is the most like surreptitiously Nazi coded route you could take.
• 44:02
  Lowtax: I mean it is Argentina.
• 44:05
  George: Oh God.
• 44:08
  Propter: So yeah, so too early to say whether or not we actually have a problem. But if we do have human to human hantavirus transmission out there, this is the kind of thing that could be a real problem. World health wise. World Health Organization is on it and they're doing contact tracing in South Africa. So this is not explicit.
• 44:25
  George: They made it all the way to South Africa before they had. They discovered the issue or I don't
• 44:30
  Propter: know when they discovered the issue. They got people to the hospital in South Africa. So this thing, this thing has like a three week incubation time.
• 44:39
  George: Oh, that makes more sense because I was gonna say like that's gotta be a pretty long trip. Like, I mean, how are they only getting sick once they're in south. But yeah, if three week, you know,
• 44:48
  Propter: I mean they also like they stopped, I think they stopped in Antarctica. They stopped in the Falklands, like did. There were a few stops here. So it's not impossible that they got hantavirus somewhere else.
• 44:59
  Lowtax: Oh, is this, is this Argentina's latest plot to get the Falklands back?
• 45:07
  Propter: Just take it right back to being uninhabited.
• 45:09
  Lowtax: Right. Just going to kill all the Brits with antivirus.
• 45:14
  George: I would not put it past that insane, insane man to pull some sort of hantavirus based or hantavirus based biological warfare on the Falklands of all places.
• 45:28
  Propter: Anyway, so insofar as you were worried about zoonotic diseases or zoonotic pandemics, tick that up like one notch because we've got something bad happening that may or may not be human to human transmission, but is definitely confirmed viral cases in a place where they should not be. And also it's killing people.
• 45:48
  Lowtax: Well, the good news is we have professionals in charge.
• 45:51
  Propter: Yeah.
• 45:53
  Lowtax: Experienced and well versed in all of this.
• 45:56
  George: Yeah. The US is funding the WHO and the CDC is in good hands. So no questions. No, no questions there. Hey. And you know, we've also got this really, really popular, you know, widely used now. Well supported by government technology that can quickly vaccinate against emergent diseases.
• 46:18
  Ed: So.
• 46:19
  Lowtax: God, I just had a cursed thought of there being a pandemic and you having to watch press conferences with rfk like we used to have to do with fauci.
• 46:29
  Propter: Oh my God.
• 46:29
  Lowtax: Oh boy, could you imagine having to endure that?
• 46:35
  Propter: It's zoonotic too. Like he's going to suggest we eat the rats or something.
• 46:39
  George: He's going to. Go get your local roadkill and fry it up and that will provide you immunity. I Need more hgh.
• 47:19
  Propter: We also had some exciting things happen up in, up in the Northeast in recent days. Do we want to talk about that?
• 47:27
  George: Yeah, we're going to talk about.
• 47:28
  Ed: No, no, we don't.
• 47:29
  George: Ed, you don't have to enjoy it. And we can edit, we can edit it out. Whatever you say.
• 47:35
  Ed: I'm just here so I don't get fined.
• 47:37
  George: We're arrested. Janet Mills dropped out in. I mean, I think the writing from the polling was pretty clearly on the wall for the past two months. That her candidacy, Governor Janet Mills of Maine, candidacy for the Senate race there was dead in the water and it's now dead. No longer in the water. It's under it. It's gone.
• 48:00
  Propter: So Graham Platner is the presumptive Democratic Senate nominee for the great state of Maine. There are, there is technically one other person on the Democratic ballot plus a write in candidate who is running that we won't spend a lot of time on. But Platner is going to win. There's no reason to think anybody besides Platner is going to win. He was at 60 in the last poll and I would be surprised if anybody even pulls it again before the primary. What this does mean is that Platner is not going to have to debate before he debates Collins in the GE since he didn't debate Mills. So that's going to be something to look forward to if you're a politics sicko. I would say. I would say that we have. This has prompted some opinions from some people on the nature of harm reduction and the nature of voting and what you should do with your vote if you are in a position to vote in the main ge.
• 49:02
  Lowtax: I will not be voting for him.
• 49:04
  George: I won't either.
• 49:06
  Propter: I won't be voting for him. But if I lived in Maine I would because I think that harm reduction is important here. I have deep misgivings about Platner. I think at minimum he's a. And at maximum he's a pretty bad guy. But you take the best option you have with the information you have. And right now with the information I have, I think that Platner would be a better senator for Maine and for the country than Susan Collins would. Because we know who Susan Collins is. She's one of the best Republicans in the Senate. But the nature of party polarization in this day and age is that the best Republican is worse than the worst Democrat. And it kind of sucks that that's true.
• 49:48
  George: It's frustrating at the very least.
• 49:50
  Propter: Yeah, that hasn't always been true. I think that has been true for all of our adult lives, at least. But historically there have been people from one party who sit inside of the other party's spread on a variety of issues there. Even though. Even though, for instance, during the civil rights era, the Democratic Party was clearly better than the Republican Party on civil rights. There were Republicans who were good on civil rights and there were Democrats who are really bad on civil rights. There was just more spread in terms of ideological commitments in both parties back then than there is now. I don't know if it makes sense to mourn that because I don't think we're getting back there anytime soon. We've moved towards strict party sorting and it's hard to see how that breaks, barring some issues becoming extremely salient, that for whatever reason, split party coalitions. So far, every splitter type issue we've directed at the current matrix of party coalitions has just resulted in it being absorbed by one party or the other and one party moving towards that issue. And we're basically still just straight sorted. Hard to see how that changes.
• 51:12
  George: It will change at some point. It will. Because politics is a dynamic equilibrium game. There is no such thing as a permanent equilibrium in the great game to decide who gets to do what at the, you know, point of a sword of government. But I don't think it's changing anytime soon.
• 51:33
  Propter: So, you know, what do you do if your options on the ballot are people that you don't care for?
• 51:41
  Ed: And I suggest moving. I did. It's worked out fantastic for me.
• 51:48
  Lowtax: I think if it were, I'm with Proctor, I think if I had a vote, I would not be thrilled about this, but I would vote for Oyster Boy.
• 52:00
  George: I would, too.
• 52:01
  Lowtax: He's. He's better than. It helps that I'm a college football fan. Like, I've had to deal with morally compromised people watching them parade around my entire life.
• 52:13
  Ed: So, yeah, see, I've been a fan of good sports. So, like, you know, like those upstanding Canadians.
• 52:18
  Lowtax: Yeah, yeah. Outstanding hockey players.
• 52:21
  Ed: Yeah. I mean, look, actually, I'm not going to say that on air. I'll leave that one to. To the philosophers to wonder about. I'm just going to say, okay, yeah, move out of Maine. It's good.
• 52:41
  George: If you think there's teeth here. If you think about a hypothetical future where he's the 52nd or 51st Democrat in the Senate, he has to be so, so bad on so much to be worse than that body remaining in Republican control. He has to be so bad. I honestly think I like, I would crawl across broken glass to vote against Fetterman in a Pennsylvania Senate primary if I lived in Pennsylvania. I think the guy is terrible. He is right in line with the party on a huge number of issues and as bad as just awful as he is on Israel and Palestine and as big of a troll as he is, and as many Trump nominees as he's voted with.
• 53:34
  Lowtax: On immigration.
• 53:35
  George: On immigration, he's not great. But there's a ton of issues where he's right in line with the party and he is therefore better than any Republican alternative. It doesn't matter. In a primary, it's a completely different story. Go to hell. Like, I, like, I don't care if it's Conor Lamb. And many apologies to Connor Lamb for myself and many other people, but, you know, like, we were not familiar with your game.
• 54:01
  Lowtax: Connor Lamb, Yes.
• 54:03
  George: Just
• 54:05
  Propter: to be fair, Conor William hasn't proved that he's good yet. We've just proved that Fetterman's really bad.
• 54:11
  George: Right, whatever.
• 54:15
  Propter: Of all Pennsylvania politicians right now.
• 54:18
  George: Like, so. But the point is, I'm using Fetterman as the example because he is, I think right now the worst Democrat in the Senate. Yes, clearly, very clearly the worst Democrat in the Senate. He is way, way better than any Republican alternative. And I think there is a really good case to be made that Platner will be much less bad than Fetterman. I think, I mean, I think there are a bunch of things that Platner says and way he way stuff he cares about, clearly, that are very closely aligned with my politics, even though there are things about him that could not be less aligned with my politics. But I, I am very skeptical. He is going to be as big of a troll and as brain damaged, literally, as Fetterman. And so if he's not even worse than the worst Democrat in the Senate right now, you gotta be nuts to vote for a Republican instead of him.
• 55:10
  Ed: Him.
• 55:10
  George: You got to be nuts. And that is really, I do think that is a wrecker type perspective. I, I really, I really do. And I, I just want to reiterate, like in a primary and like, it, like, this guy sucks. Like Platner sucks. But it's not that he sucks, it's
• 55:27
  Ed: that he's a Nazi.
• 55:28
  George: Yes, okay, sure. Yes.
• 55:31
  Ed: That's what it is. Like, we're not talking. We're, we are off of, we are off of the, the map here. Like, the dude was running around for 20 years with a fucking Totenkoff.
• 55:43
  Propter: That's, that's the question. Is he A Nazi. Right.
• 55:45
  Ed: Like he's either a Nazi or he's incredibly friendly with them. I don't see a second option or a third option. I can count. Right. Like this is, this is the problem here is that, yeah, like he's. If he's elected, is he probably better than Fetterman? Yes. He's also still either a Nazi or okay with them. And like, that is a stain that you cannot erase. And that is a real problem. Like, it's not just him, although him and his, his dipshit friends. It's the way that like Liz Warren is out there scraping for this guy. Yeah, right. Like the Democratic establishment has stained themselves with this guy. You're probably right that you have to vote for him if you live up there, because sure, he' probably not as bad, but we are going to be answering for this guy for the rest of our fucking lives.
• 56:39
  Propter: I think Ed's probably right about that. And yeah, I think so. I would say that if you are a Democratic politician from other states, you should be a little careful on this one.
• 56:48
  Ed: Yeah. Like I am never voting for Liz Warren again.
• 56:51
  George: I'm absolutely not. I think it is very different from him winning the primary, winning and being the general election, general election candidate and Liz Warren coming out and throwing her weight behind him at that point. I think that is as proper reduction. But I think Liz Warren and also Bernie Sanders. Oh yeah, Bernie Sanders came out guns blazing for this guy. And I, look, I am, brother Bernie has had a huge influence on my politics. He, I think, is one of the best hearted people in American politics. I think some of his report, his supporters are annoying and shitheads, but I think he is the real McCoy and there are not that many people like him in American politics. And it's rare that I say, you know, Bernie, you really screwed this one up in a way that matters. Like it really matters. And he screwed this one up.
• 57:45
  Lowtax: I mean, Bernie has never been all that great about picking the people around him.
• 57:50
  George: I totally agree with that. That is his biggest failing as a politician. I completely agree with that. But. But this goes a little beyond that. This is not picking an advisor or picking someone who's not. This is a public political coalition and welcoming with open arms someone who is. Maybe he's not actually a Nazi, but boy, was he close with a lot of them. So, yeah, yeah, it's.
• 58:16
  Ed: I think everyone who endorsed him before the primary was over should never be voted for again. Every single fucking one of them, without exception.
• 58:26
  George: Yeah, I think that's a fair take.
• 58:28
  Lowtax: Yeah, I Think so too. I think it's.
• 58:29
  Ed: I am going to max whoever runs against Liz Warren. And you all can keep me honest on that.
• 58:36
  Lowtax: Yeah, I mean, look, Schumer and Jeffries, I think, came out and endorsed him once Mills dropped out. That I like. It is what it is. I don't have a problem with that, but.
• 58:48
  Ed: Yeah, but they should be answering for it for the rest of their lives. It is different, but.
• 58:52
  Lowtax: Yeah, sure, but it is different. It's different from getting behind this guy
• 58:59
  Propter: while there's still a contested primary.
• 59:01
  Ed: Right, that, that's a little galling too. Yeah.
• 59:05
  Propter: Anyway, unfortunately, you know, blue, blue no matter who, means that sometimes you got to eat. And yeah, this is one of those cases. It sucks. It sucks when you lose one. It sucks when you have to compromise. It sucks when you have to betray one set of principles in the name of another set of principles. And you know, it doesn't just suck for me, it sucks for other people. It sucks for other people with principles who are different than mine. It sucks to have to deal with the difference between your vote as an instrument and your vote as a part of your self image. Sometimes your vote only matters in terms of what you believe it says about you because it's not going to make any difference to the actual politics of power. Other times your vote matters a lot because it's an instrumentality. It's something that does directly affect what can be millions, what can be billions of people's lives.
• 1:00:12
  Lowtax: On the upside, our other Senate candidates I think are pretty good. I like Roy Cupper. I like.
• 1:00:19
  Ed: I don't like how much I have to hear about him though on this podcast. We have some real North Carolina biases.
• 1:00:25
  George: I didn't bring him up. I don't know why you're getting mad at me. Outer Florida was the one that brought him into this conversation.
• 1:00:31
  Ed: I believe the term.
• 1:00:32
  Lowtax: We have spent probably seven full episodes worth of discourse on Maine, which you all brought up.
• 1:00:41
  Propter: We did. We did.
• 1:00:43
  George: Ed did specifically not want to talk about this and we told him we had.
• 1:00:45
  Ed: Yeah, I want. I want to fucking disown the state. I'm glad I left. I find this to be embarrassing, like at a mortal level. Well, I mean, I think this state should be returned to Massachusetts. It is not allowed to self govern.
• 1:01:05
  Lowtax: And then we find out that Tucker lives in Maine.
• 1:01:08
  Ed: All right, I'm upgrading this to full nuclear exchange.
• 1:01:13
  George: One sided, of course. Maine doesn't have any nukes, does it?
• 1:01:16
  Ed: No, I don't probably.
• 1:01:18
  Propter: They build submarines up there, right?
• 1:01:21
  George: I think they've Got a submarine base, but they don't build them in Maine.
• 1:01:25
  Ed: I thought. Used to.
• 1:01:29
  Lowtax: I think all the important naval stuff is down in Georgia.
• 1:01:33
  George: Yeah, in Georgia.
• 1:01:35
  Lowtax: Yeah. What the hell is that base called? Where we keep our nuclear subs.
• 1:01:41
  George: Isn't that in Hampton Roads?
• 1:01:44
  Lowtax: No, I don't think so. It's like a king. King something.
• 1:01:51
  Propter: I should know this. I don't know this.
• 1:01:52
  Ed: Oh, no. BiW did destroyers in World War II.
• 1:01:56
  George: Okay.
• 1:01:56
  Lowtax: All right.
• 1:01:58
  George: Knowledge about the military industrial complex with respect to the Eastern seaboard is now on full display. And no one's happy about that.
• 1:02:07
  Lowtax: Our other Senate candidates are good, though.
• 1:02:09
  Propter: Our other Senate. Other Senate candidates are. Are good.
• 1:02:12
  Lowtax: We've got Cupper, we've got Talarico, we've got Salmon lady in Alaska. We've got Brown. Good old Sherrod is coming back in Ohio. Do you.
• 1:02:25
  George: What's the. Where are we at on pricing for that right now? And not. Not that. Not that like prediction market pricing is necessarily anything more than consensus, but, you know, I still think the most.
• 1:02:38
  Lowtax: I haven't seen any prediction markets. The polling looked like it was probably 50.
• 1:02:44
  Propter: 50, 55, 45 polymarket 60 got at 60. 40 dam right now.
• 1:02:50
  George: Yeah.
• 1:02:50
  Lowtax: Oh, wow.
• 1:02:51
  George: Cal. She. I'm actually surprised it's that high. I'm surprised it's that high and that Senate Dem isn't higher, honestly.
• 1:03:02
  Lowtax: Do we know who's going to get the nomination in Iowa?
• 1:03:06
  George: We do not, actually, we don't. But the candidate that announced this week is really, really good. He's a pastor from. Oh, cool. Well, good in terms of like super electable. He's a pastor for a local string of churches and has really good name recognition and, you know, just seems like a really high quality candidate that sees an opening and this is my best chance to win this seat for the next ten years or longer. And let's. Let's go for it.
• 1:03:38
  Propter: I think that's. I think that's. I think that's Indiana you're talking about, not Iowa. For what it's worth.
• 1:03:43
  George: Give me a second
• 1:03:47
  Propter: because I.
• 1:03:48
  Ed: We're really outing ourselves on middle of the country geography. Nope.
• 1:03:54
  George: Hold on.
• 1:03:55
  Lowtax: One of those stupid corn st. You're talking.
• 1:03:57
  Propter: You're talking about David Green, right?
• 1:03:59
  George: Give me a second. This is the problem with using your Blue sky feed as external memory marker. I'm talking about Adam Hamilton and he's from Kansas.
• 1:04:10
  Propter: Kansas.
• 1:04:12
  George: Can we just edit this? Like, please, whoever's doing that, I can't do the edit this week.
• 1:04:17
  Lowtax: So the other, other. Other corn states.
• 1:04:20
  George: I Mean, confusing. Iowa and Kansas is, for the record, very forgivable territory. Like, come on, they're both squares full of corn.
• 1:04:27
  Ed: Like, if you would swap to Iowa and Ohio, we would have a problem. But honestly, once you get out to, Once you get out to Iowa and, like, points west slash south. Look, I'm just saying other states need to have their governance revoked as well.
• 1:04:44
  Lowtax: Kansas versus Nebraska versus one of the Dakotas, sure, but Kansas versus Iowa, I'm not giving you that one.
• 1:04:53
  Ed: Look, I, I tried to help George. He's leaving in shame now, but, you know.
• 1:05:01
  Propter: All right, so, anyway, we have exciting Midwest candidates. We're not quite sure where they are. We're a little hazy on who they are, but we're excited about them. That's actually true. I, I, I, I look forward to finding out more about our Midwest candidates as we determine which of them have a real shot.
• 1:05:17
  Ed: Yeah, no, that's about right.
• 1:05:19
  Lowtax: I mean, with where Trump's approval is right now, like, it's conceivable that you're competitive in Kansas.
• 1:05:26
  Propter: Yeah.
• 1:05:27
  Lowtax: Kansas is one of those states where if things line up just right, you get the right environment and stuff, you can, you can pull it off.
• 1:05:35
  Propter: It's, I mean, the timing is right for people to be mad about fertilizer prices, right?
• 1:05:40
  Lowtax: Yes.
• 1:05:41
  Propter: So here's hoping.
• 1:05:45
  Lowtax: And Alaska, again, all the salmon you can farm, just, just vote for, Just vote for Peltola.
• 1:05:57
  Propter: For those who may not be in online left politics quite so much, Mary Peltola is the former representative from Alaska who ran on the slogan pro choice, pro fish.
• 1:06:13
  Ed: And did the fish get a choice?
• 1:06:16
  Propter: I hope not, man. So, you know, she's, she's great. We're, we're big Palto fans on this podcast. She would be intensely frustrating if she lived in a state that wasn't Alaska. But because she lives in Alaska, we love her.
• 1:06:32
  Lowtax: I don't know.
• 1:06:32
  Ed: We.
• 1:06:32
  Lowtax: If she lived in Idaho, we'd build her a statue.
• 1:06:35
  Propter: Yeah, that's true. Actually. She's in, she's in a red state. Put it this way. Like, she and Marie Glencores voted the same kind of a lot. So, like, however you feel about MGP maps, pretty well to how you should feel about Peltola. Except that Paltola lives in a place that, you know,
• 1:07:01
  Lowtax: also, Peltola doesn't sell meth.
• 1:07:04
  Propter: Well, there's that. But she probably does want to shoot sea lions.
• 1:07:09
  Lowtax: I'm kind of with her on the sea lion thing.
• 1:07:11
  Propter: She's right about the sea lions.
• 1:07:12
  Lowtax: I'm not, I'm Not a big fan of the Sea Lions
• 1:07:16
  Ed: or the Detroit Lions.
• 1:07:20
  George: Why?
• 1:07:20
  Lowtax: You can't hate on the Detroit Lions. It's Detroit. Like, you can't hate on it. Everybody. The Republicans all hate on Detroit and stuff. Leave. Leave poor Detroit alone.
• 1:07:33
  Ed: But. But this is. This isn't Detroit. This is their football team.
• 1:07:38
  Lowtax: Yeah, I know, but think of all the repulsive teams in that division, though. Like, you've got the packers, the Bears, I guess the Vikings are okay, but are they.
• 1:07:53
  Propter: No, that division's cursed.
• 1:07:55
  Lowtax: No, not really. I would have handed it to him if they hadn't got Brett Favre.
• 1:08:01
  Ed: I was gonna say, like, that seems damning. Not as damning as certain other things we've discussed on this podcast today. Yeah, but, you know, pretty bad.
• 1:08:14
  Lowtax: I don't know. I bet Brett Favre's got worse tattoos than a toten cop.
• 1:08:21
  Propter: He hasn't exactly covered himself in glory in recent years. Mostly welfare fraud.
• 1:08:26
  Lowtax: Yeah, Yeah, I was gonna say. I don't. At least Graham Platner is not stealing Medicaid money from poor people and stuff.
• 1:08:33
  Ed: Are you sure?
• 1:08:35
  Propter: Well, he's not in office yet.
• 1:08:39
  Ed: Have we checked on Covid stimulus? Sorry, was that too true?
• 1:08:47
  Propter: No, I mean, that's. That's a fair question. We probably should have done a little bit more oppo on the guy with Nazi tattoos.
• 1:08:52
  Ed: I am. I am one like one.
• 1:08:56
  George: Jesus.
• 1:08:57
  Ed: People in Maine should vote for him. Sigh. But I really do wonder what the oppo is going to look like here.
• 1:09:04
  Lowtax: How is this guy a bartender in D.C. on Capitol Hill, and nobody's got fucking oppo on him?
• 1:09:15
  Ed: We're all just gesticulating over here. Nothing we're going to say on a microphone, but, you know. Yeah. Anyway, do we have anything uplifting or nice to. To talk about this week? Because I'm feeling a little cynical.
• 1:09:53
  Propter: I kind of just had that. The hantavirus thing.
• 1:10:03
  Ed: We're doing great, folks.
• 1:10:07
  George: Depending on your valence. The earnings results this week from the biggest tech companies are either fantastic news and the economy is looking like it's going to have some AI Capex rocket fuel for at least another year or two. Or really, really terrible news because you think AI is basically a devil technology and should be banned, et cetera, et cetera.
• 1:10:32
  Ed: You know, we've already read your posts on Blue Sky. You do not need to follow up again.
• 1:10:41
  George: Okay, so. All right. I got my Fed segment earlier, so we don't just.
• 1:10:45
  Ed: No, I don't mean you. I mean the people who are going to complain about the AI. Please. Okay, yeah, that did that did seem kind of mean. Yeah, that. That was at our posting. Enemies.
• 1:10:55
  Lowtax: That's not.
• 1:10:56
  George: Okay, okay, okay.
• 1:10:57
  Ed: I actually didn't read George's posts.
• 1:10:59
  Propter: Come on.
• 1:11:00
  George: Yeah, he never does. It's fine.
• 1:11:02
  Ed: I only read posts to comically misunderstand them.
• 1:11:05
  Lowtax: No, you only read my post so you can yell at me about the PlayStation 2.
• 1:11:09
  Ed: Yeah, well, you deserve to be yelled at.
• 1:11:11
  Propter: Yeah, you're very wrong on that, man.
• 1:11:14
  Lowtax: I'm sorry, what was. All the consoles that. All the consoles after the Super Nintendo and before the PS4 were trash.
• 1:11:26
  George: I. I don't have a way to evaluate that.
• 1:11:28
  Lowtax: Just trash games. Like weird polygonal.
• 1:11:32
  George: Just Halo alone, it makes that wrong.
• 1:11:35
  Lowtax: No, Halo sucks.
• 1:11:36
  Ed: Yep, he's right about that one.
• 1:11:38
  George: I'm not talking about Halo 18 or whatever the. I'm talking the original Halo Combat Evolved. Like, that game rocked. What are you talking about?
• 1:11:47
  Ed: Well, see, they called it Evolved, but the rest of us had been playing, you know, real shooters for years at that point, before the Xbox got one.
• 1:11:58
  George: Did y' all never do, like, the four on or two on two, Like LAN party? Like, what was that gulch map? That, like, red gulch map with the rocket launchers? Blood Gulch. I just hate fun. Like, I don't. I mean, this tick is so bad that, like.
• 1:12:15
  Ed: No, but I'd already been playing video
• 1:12:17
  George: game take on this podcast, which is like something that's never happened before.
• 1:12:21
  Ed: I had already been playing Unreal Tournament at this point. Like, I had been playing with like 32 player maps of. On. On games that don't require you to play with a controller. Look, first person shooters getting to consoles, it was good for the console players, sort of. It actually wasn't because Xbox Live came out and that had voice chat. That was a mistake.
• 1:12:44
  Propter: Yeah, long term, that was terrible for consoles.
• 1:12:46
  George: Yeah, I agree with, like, the health
• 1:12:49
  Ed: of the human race. But, like, I mean, it was good for what it was. It's just. It was also a game that, like, when I was in college, like, Halo 2 was the game of choice. I would play it and I'd be like, this feels really bad. Why am I winning? Right? Like, it was just. It was not built. It was the. It was an evolutionary misstep for the. The. The. The college set. That didn't make any sense. I'll try again.
• 1:13:20
  Lowtax: All of Sony slop fans are accusing me of being a Halo fanboy. I have never touched Halo in my life.
• 1:13:28
  Ed: In fairness, you sound like it.
• 1:13:30
  Lowtax: We played COD United Offensive in that time period for multiplayer. Okay, we did not play Halo I
• 1:13:39
  Ed: can, I can get behind that wasn't mine. But you know, Half Life already existed so like I was busy but oh no, oh no, your takes are about to get worse, aren't they?
• 1:13:54
  Lowtax: Oh, what about Half Life? I have no real opinion on Half Life.
• 1:13:57
  Ed: Okay. You made a face that I was like, oh no, no.
• 1:14:01
  Lowtax: My buddy got me a copy of Half Life on Steam one time and I played it for like an hour and I was like, it's fine.
• 1:14:13
  Propter: So hyperscammers.
• 1:14:15
  Ed: Yeah, yeah, let's just, let's go talk about something nice like Jeff Bezos's money hats.
• 1:14:22
  George: So two of them traded up pretty well. Google and well, Alphabet, but it's Google. Google and Amazon traded up pretty well on their earnings and Meta and Microsoft did not. They Both trade about 6% lower, 6 or 9% for Meta on the earnings reaction, 4 or 6% for Microsoft. But the key thing is Meta really doesn't have legs. Yeah, they're all continuing to raise guidance. They're all saying capex is going to keep growing and so until that changes, this is just going to keep happening. Microsoft has 680, I think is the number 686 billion 683 billion of remaining performance obligations. What that is is contracted revenue that you haven't earned yet. You've got an agreement for someone to pay you, but you can't book it as revenue until they actually pay you under accounting rules. But RPO kind of gives you an idea of, you know, what that number looks like and like a quarter of it is in the next 12 months and, or, sorry, half of it is in the next 12 months, I think. And that's just so much money. That's just so much money. Like they,
• 1:15:38
  Propter: so, so the big companies are doing really well here is kind of, is kind of the takeaway. Is that, is that fair to say?
• 1:15:45
  George: Yeah, I mean I think it's less like they're doing well and more like they continue to see signs that plowing all this money into capex is going to generate returns. And so they're going to keep doing it. And so the wheel keeps spinning. I mean eventually this will peter out, eventually they'll. But for context, I looked at a couple different things this week. I looked at the operating income, so earnings for interest in taxes for six memory companies. These are like SanDisk is one, the Korean pair or another two. There's, you know, whatever. And these six companies who, all, all these guys do, I mean it, it's. Some of them do other stuff too, but basically all they do is make random access memory.
• 1:16:28
  Ed: These are the flash companies, right?
• 1:16:29
  George: Yes.
• 1:16:30
  Ed: Or are you also doing dram?
• 1:16:31
  George: No, sorry. This is Samsung, SK Hynix, Sandisk. Okay, so the dram Micron. Yeah, DRAM companies, which is where the, the, the sandisk is.
• 1:16:40
  Ed: The flash company is always confused.
• 1:16:42
  George: Yeah, it's the publicly traded direct proxies for the memory demands of the AI capex. Boom.
• 1:16:50
  Ed: Gotcha. Okay, gotcha.
• 1:16:51
  George: Whereas like Nvidia is the processing power. Right. So those six companies are projected to have operating income equal to half a percentage point of global GDP in like two years.
• 1:17:07
  Ed: I hate that.
• 1:17:08
  George: And the thing is that forecast looks conservative when you look at how much their operating income has risen in the past 12 months. If you look at the hyperscalers. So that's Oracle, Meta, Google, Amazon, Microsoft.
• 1:17:27
  Ed: It kills me that we say Oracle is a hyperscaler now.
• 1:17:29
  George: They are. I mean they are, the numbers are there. That's the right way to think. They're the smallest, but they're there. If you look at them, are they
• 1:17:38
  Ed: within like 50% of any of the majors at this point in terms of
• 1:17:44
  George: Capex or in terms of revenue?
• 1:17:47
  Ed: Revenue?
• 1:17:48
  George: No.
• 1:17:48
  Ed: Okay, we need a word that's below hyperscaler here.
• 1:17:53
  George: Well, okay, but like in terms of big publicly traded cloud services providers, they belong in that bucket, right? Like, because, because the next, the other people you compare them to be like what, like Core Weave? I don't think that's a fair comparison either.
• 1:18:05
  Ed: I haven't thought of Core Weave in years.
• 1:18:07
  George: And yet the neoclouds are like a whole thing. So I mean I, whatever, we're down a rabbit hole here.
• 1:18:15
  Ed: If I had Oracle in like the IBM cloud bucket.
• 1:18:19
  George: Oh, IBM cloud is like tiny compared to Oracle.
• 1:18:22
  Ed: Okay, like I mean like I don't even hate Oracle cloud so like I, I'm just kind of pleasantly surprised.
• 1:18:28
  George: So the thing is like Oracle cloud is tiny but then they've got this, this Stargate business that is like oh,
• 1:18:35
  Ed: are they folding that in there?
• 1:18:37
  George: It's all. That is a cloud services business. I mean you know, whether you put it in the same bucket in terms of scale as Azure or Google Cloud or aws. I mean fine, it's, it's not, but
• 1:18:48
  Ed: it is still a, it makes total sense. Azure plays games with it too.
• 1:18:52
  George: It's a 12 figure revenue business in the long run is the, is the kind of thinking here. And okay, whatever. Those five companies once again because we got sidetracked by my co host several times. Oracle, Meta, Google, Amazon, Microsoft, those five companies, their combined CapEx forecast through it basically rises through the rest of this year into 2027 and starts to peak out in late 2027, early 2028. At that point it will be roughly equivalent to 2% of US GDP. Two and a half percent of US GDP. It's not all spending in the US so like you can't literally compare them. Like it's not, it's not, it's, you know, these are global companies. They have capex in other places beyond the US so like it's not fair to say all of that capex is driving US activity but like just as like a sort of scaling exercise, right? Like, like to kind of get appreciation for the numbers. That is so much spending and like, like it's, it's going to keep rising and then plateau at 2 1/2 percent. It's not projected to drop after 2028, at least in, you know, maybe way far out years. It will. So yeah, I'm the, the whole AI thing is not, it's not slowing down. And I really like, I like there are downsides to this technology, we've talked about them. But if you're a person that thinks, oh well, like it's just not useful and there's no market for it and like it's all like a plagiarism machine and like you shouldn't be using it and like whatever, whatever, you're perfectly entitled to that opinion and like principled takes are fine, but you are going to be left behind really fast. Like you're. This, this.
• 1:20:34
  Ed: It's here and I am sorry. Go ahead. Proper.
• 1:20:39
  Propter: I was just like, to be fair, one of the nice things about AI as a technology is that it's pretty good at onboarding you when you're ready to get on board with it. Like, like this is not, this is not something where you know you're necessarily going to need to spend a lot of time learning things outside your skill set to be able to make good use of AI because it talks to you in the language you speak.
• 1:20:58
  George: Especially if you're the type of person that is good with words to begin with.
• 1:21:02
  Ed: Yeah, no kidding. This stuff is designed for people who
• 1:21:05
  George: write the age of the word sellers nigh.
• 1:21:09
  Ed: We're on to the word rotators. George, keep up. This is a uninformed take, but that's never stopped me before. The amount that all of these companies are spending and like the amount of revenue that they're all forecasting really seems worrisome in a way to me. Like I, I do not know These companies from the financial side, but I do know from a product side and I think a lot of these companies are going to start getting their tickets punched in the next year like really hard by the China models. China has figured some shit out in terms of like qualitatively better stuff and there are always going to be some moats around like data retention, security concerns. Misguided though many of them are. But like this is, this is why like OpenAI and Anthropic in particular are like trying to move up and Microsoft seems best poised to do that in terms of like ecosystem. They've actually partnered with Anthropic on like a cowork for Copilot, which I find very strange. But like just on a cost basis they're not going to be able to keep charging the money that they're charging. Not to the audiences at the sizes that they are. Like Xiaomi has a frankly bonkers model in MIMO 2.5 and it costs 3$3 per megatoken output while Claude is 25 and GPT 5.5 is 30. Right. Like this is getting really weird in ways that I don't know how they deal with.
• 1:22:46
  Propter: So this is, I mean I think this is at least a little bit complicated because there are certain things that you use a whole lot of tokens for and there are certain things that you don't use a whole lot of tokens for.
• 1:22:59
  Ed: Well you also use a whole lot of tokens in different ways depending on the model. So it hits a bit weirder than that.
• 1:23:03
  Propter: Fair?
• 1:23:03
  Ed: Fair.
• 1:23:04
  Propter: So like some of this is what kind of applications you're trying to do. If you're trying to do a lot of writing of code or something else that is really taxing in terms of the actual throughput on the model, then that number per million tokens is really important to you. If you're trying to do something where you need good surfaces for accessibility so that it's easy to onboard your company to the model, but you're not going to use all that much of it, then somebody who has more approachable surfaces but has a higher per token cost might be a better fit for you.
• 1:23:42
  Ed: But you're talking like 8 to 10x.
• 1:23:46
  Propter: Yeah, but, but like I mean this isn't, this isn't the only place where you see business to business services that are, that are, that are differentiated along.
• 1:23:55
  Ed: Sir, I work at the phone company.
• 1:23:56
  Propter: I understand.
• 1:23:57
  Ed: Yeah, like it it and also like taking into account. But how many people are going to want to be dealing with the American tech companies after that's, that's a different question.
• 1:24:10
  Lowtax: Right.
• 1:24:10
  Propter: Like I mean if it turns out, if it turns out that the Chinese companies are easier to deal with and also are much better on price, then yeah, they're, they're going to eat our lunch.
• 1:24:18
  Ed: Like it's interesting because like the Googles of the world have spent a very, very long time building out systems that are very difficult for them to support. Right. Like there are many Googles in lots of places and people are even getting skittish about that. The US is you know, not the only market out there which we are likely to find out to our severe discomfort. But like it just, I struggle to see a pencil out because like I don't see. I agree with you right now in terms of the like rp, is a company going to come on and quote unquote, do AI and spend a ton of tokens right now? Probably not, but the way that all of these things are designed is they want you to spend more tokens over time time. And so you start getting into either these are captive plays or you're gonna start looking for places to, to cut that budget. I can tell you that where I am, they very recently capped everyone's clod spend at $50 a week. Which that's getting to the caveman, banging two rocks together stage of things. And, and like that's purely just on the cash shock. Right. Like I, and, and like there are good reasons to be using, you know, US based companies for that stuff, but that's still pretty wild.
• 1:25:39
  Propter: Yeah. And in so far as that's, insofar as that's a major throttle point, that's, that's going to be a problem and it may be that eventually the pricing model is going to have to move. The multi billion, maybe multi trillion dollar question here is to what extent does utility to the buyer scale with tokens? And for some use cases it's a very direct scaling. It is clear that cost to the provider company scales with tokens. That's unambiguous.
• 1:26:08
  George: Right?
• 1:26:08
  Ed: I mean I think it's actually sublinear for them but that's because of like resource batching and weird shit.
• 1:26:14
  Propter: Sure, maybe it's a muted scaling, but still basically the more tokens you're using, the more it costs anthropic to serve, to serve you those tokens. So I think that in terms of pricing plays we're going to see a lot of interesting things about what companies are trying to do to fit one to the other. Some of the applications that one can imagine for this, don't necessarily use a lot of tokens either in the initial state or in the steady state. And some use a gigantic amount. We talked about this before. Like I'm, I'm pretty bullish on the utility of the kind of model that you can run on a box in your basement in a year or two. It's easy to imagine that consumer grade hardware is going to be able to run a model that you can, you can get for free, that is going to be able to do what you want it to do for most purposes. It's not going to be the frontier model. It's not going to be as smart as the smartest thing out there, but it's going to be able to do the kinds of things that you can do.
• 1:27:10
  Ed: Right now with Claude, you're, you're too soon. Like, like the, the ramp. I, I think you're, you're right about it, but it's like 2030 just because of like being able to get access to.
• 1:27:21
  Propter: So maybe it's, maybe it's four years, maybe it's six years. Like it doesn't really change the.
• 1:27:25
  Lowtax: Under the.
• 1:27:26
  Propter: Under.
• 1:27:26
  Ed: I agree in terms of like individual things, I'm a lot less certain about like you start getting into things like data homing and you know, where your compute lives. For companies that starts being weird. Right? Like one of the things that I am waiting for Apple to get their ducks in a row and release a Mac studio because I want to throw 256 gigs of RAM at things, mostly models, but like that's going to be a rather tough pill to swallow for a lot of people. And like centralized governance and stuff is really the basically making models work with, with working together is the hard part right now. Which is why, you know, cowork exists. It's why OpenAI is chasing up with Codex. Why again, Microsoft might actually win because they're here first. The thing that's hard is that the moats are really shallow now because like you can go build, you can go build a reasonable knockoff of these things as long as you can get enough GPUs. And that's a really weird spot to be in. I'm not saying that they won't make money or that they're all going to disappear or anything, but it just, it feels weird.
• 1:28:36
  Propter: Yeah, it feels like there's a lot of fragility there. And at the same time that fragility is being coupled with like George said, this, this gigantic capex. And you know, to some extent the capex gives you Some reduction in that fragility. You're buying things that are presumably not purpose built to the extent that they're not going to be useful for somebody even in the event.
• 1:29:00
  Ed: No, they're very general purpose.
• 1:29:02
  Propter: That the case you're building them for doesn't work out exactly like you thought. So like, you know, those are, those are real assets that exist in a way that is less ephemeral than maybe we're used to talking about with tech company assets. It's not dependent on market share. It's, it's, it's.
• 1:29:20
  George: Well, there's depreciation and like. So the, the canonical example for tech company assets that were generally useful would be the fiber build out in the late 1990s, early 2000s. Right. Which, you know, that doesn't really depreciate. I mean it does, but like not fast. Right. Fiber lasts decades. These GPUs and the powered shells around them are. The powered shells will last 30 plus years. The GPUs will and memory, etc. Like the actual technological contents of the again, powered shell being like the transformers and cooling and roof and et cetera,
• 1:30:00
  Ed: everything down to the rack, everything except
• 1:30:02
  George: for the racks inside. Because once you have that powered shell, like you can put anything inside, technologically speaking. You can put whatever combination of GPU, CPUs, data systems that you want. The powered shell is not the most expensive part, but is sort of the hardest to build because you have to deal with the political reality of building a big thing somewhere. Then once you have that thing built, you can put almost anything in it. And I think that's where, yeah, like that computer right now, today, anybody that tells you that they think that there's no actual demand for all these data centers, they don't know what they're talking about. They are woefully hilariously misinformed. Now will eventually there be too much powered shell and all the technology inside built relative to what's actually needed in a world where you can get a good enough model on your home PC essentially. Right, as Property was talking about earlier. Oh, like for sure. I think that, I think, I think we'll get there. It's a long way away right now. And that's sort of where it comes back to these earnings calls this week is like they're still raising capex guidance. They're still reporting significant acceleration in demand sequentially. Right. Like it's still accelerating. And until it, until the at least second derivative goes negative, which is not happening, you need the second and first derivative of demand for this stuff to change and neither is happening. Both are positive still. Yeah, it's going to keep rolling along.
• 1:31:40
  Propter: Just to expand a little on that thought from earlier. Just because I think you're going to be able to run a model at home, that doesn't mean that's going to be the most efficient way to run the model. No, the scalers are always going to have the advantage there. If you're doing it, if you're doing billions and billions of tokens, the scalers are going to be able to produce that for cheaper than you're going to be able to produce it at home because they've got optimized hardware, because they can do load balancing.
• 1:32:00
  Ed: Even if you only want 20, it's going to be cheaper and more efficient to run it. Right.
• 1:32:04
  Propter: But what it means if you can do it at home is that the scalers are not going to be able to capture monopoly rents.
• 1:32:11
  George: Correct?
• 1:32:11
  Ed: Yes, yes.
• 1:32:14
  George: If there's a reason to do it at home or internal to your organization, you can think about insourcing at the organization level as opposed to a consumer level. But if there's a reason to do it, maybe you don't want, for instance, your inputs that you're putting into the model. So whatever you're typing into Claude or into ChatGPT or whatever, if you don't want the model owners to have that data, then you have to run your own instance somewhere. Right. And you can totally do that. Like that's totally doable if there's a big enough market for that, for instance. Well, now you're competing on something other than price. Right. Because like you got that trade off to make. Or if you want sovereignty over whatever you're working on for your own reasons, whether it's privacy or just a sense of I don't want another company, a company profiting off me or whatever, whatever
• 1:33:07
  Ed: it is,
• 1:33:09
  George: then you're not competing on price and that hurts total demand for all those powered shells and the technology in them. But again, like that is the place we get to where that is measurably eating into demand for those services. And those services are overbuilt relative to the demand. We're so far from that. Still, it will happen for sure. I'm not saying they will overshoot. Eventually they will, yes, I can say with 100% certainty they, they will overshoot. It's just not going to happen in a couple quarters. It's probably going to take longer than
• 1:33:43
  Ed: the real threat I do think is like non US Companies. Right. Like the, the hardest part for so Zai has been getting crushed because they literally can't get enough GPUs, right? They have a Cloud 4 point like Sonnet 4.5 ish model that's quite good. Prices are too low, honestly. And they, their biggest problem is they cannot serve enough, right? Like they can't. They, they cannot satisfy demand. Meanwhile, Xiaomi comes out of literally nowhere and is just like. By the way, here's one that is like Opus 4.7 quality, maybe a little
• 1:34:17
  George: lower because it was distilled from Opus 4.7.
• 1:34:20
  Ed: There is, well, this is slightly older, so it's probably distilled from like 4.5. But also like they're all, they're. They have their own special sauce too, right? Like it's not. These things are not purely copying. With some exceptions, I think Minimax is basically just a copy. But like Minimax is also costing ten bucks a month for like six to eight times what Claude gives you on a subscription. And what Claude gives you on a subscription is incredibly good compared to the per token costs.
• 1:34:47
  George: Like, it's crazy how much I like. I was just looking at my sessions today. So I did two things today. I built a dashboard for accessing frbus, which is the work course macroeconomic model for the Federal Reserve. They have a Python implementation of it. So I like to work in Python because everything I do is in Python. When I use Claude code, just because it's what I'm familiar with to some extent. I don't actually code particularly well.
• 1:35:15
  Ed: It's fine.
• 1:35:16
  George: I did write some Python before any of this came along, so I'm not completely lost when it comes to the actual code itself. Okay. Anyhow, so I implemented a model, or, sorry, a thing based on that. My session today was like, I don't know, half an hour time on the API. I use 20% of my daily limit. I'm under half of my limit for the week. I also built a dashboard that queries the Oura Ring API because my wife is an Oura Ring user. I built a Woop API dashboard because I'm a whoop whoop user. And I just said, hey Claude, why don't we take a look at this. Half an hour on the API, six bucks worth of compute used on Sonnet 1.7K token input, 119.1 token output, 1550 lines of code.
• 1:36:15
  Ed: You got to get those prompt sizes up. It's too, too small.
• 1:36:18
  Propter: Prompt too small.
• 1:36:19
  George: But that's the thing. Like I don't for what I'm using it for.
• 1:36:23
  Ed: No, I get It, Right.
• 1:36:24
  George: It's, it's, it's trivial and like, I,
• 1:36:29
  Ed: I just, I'm just saying those are rookie numbers.
• 1:36:33
  George: Low Tax is taking a nap. He's literally got his hat over his face.
• 1:36:40
  Lowtax: This is the worst ever.
• 1:36:42
  Ed: Yeah, it's going to get worse. George is spending $6. I'm over here. I have a session that has 19 hours of API time.
• 1:36:52
  George: Oh God.
• 1:36:53
  Ed: Wall clock time of one day, 14 hours. If I was paying API rates, this would be $1260. I pay $200 a month now. If I was at work, I would be paying $1,260 for this. But yeah, like, like, like that, that's where they're making their money is the, they're, they're definitely not losing money on me. But like the, the fact that that same workload, even if you assume it was like 30% less efficient if it was on Xiaomi, like because their models are really good, that's what, $900 difference? Yeah, for a substantial amount of work. Like, I'm building a cloud code replacement. This is just, this is not the longest run I've done in the last two weeks. And it's just the, the money is going to catch up for folks who are not either using it institutionally in like the, the, the copilot sense or people who are not like locked for reasons to. I was going to say US based models, but really it's just non China models more than anything. Our IP to Mistral, we used to think a French model would work.
• 1:38:08
  Propter: I mean, this difference speaks to what I'm saying though, right? I mean, think about what George wants as a customer versus what Ed wants as a customer.
• 1:38:15
  Ed: It's really stupid that they're trying to charge us the same way.
• 1:38:18
  Propter: George fundamentally doesn't care, at least for this purpose, about what it costs to run 100 million tokens through something. Because that's not what he's doing. He's building a thing that takes maybe 500,000 tokens and then he's done building it and then he goes and does something else that isn't that.
• 1:38:39
  Ed: Whereas I'm over here, like no, run 10 review passes, you are not allowed to stop until it comes back completely clean.
• 1:38:47
  Propter: My point is that it's not, it's not clear that one of these is a better way to use AI than the other one necessarily. Right. Like, you're both getting the thing that you want out of the AI. You're both satisfied customers.
• 1:38:58
  Ed: I'm, I'm not. That's why I'm Building a cloud code
• 1:39:01
  Propter: replacement, deeply dissatisfied customer. But. But like that kind of speaks to the divergence in the pricing models and also is why I think some of the subscription stuff might, might be better for the. The AI companies than. Than is generally being acknowledged.
• 1:39:18
  Ed: Because they're totally fine. Ed Zetron is full of shit. They're totally fine.
• 1:39:22
  George: Just to close this out, we will get a look at how fine they are because both Open AI and Anthropic are going to IPO this year. So that will come with real disclosure. Like real. Can't avoid it.
• 1:39:34
  Ed: What are. I can't wait. Because show me is right there. Right? Like they're a big boy company. They're. They are one of the big boy companies and they are charging three bucks a million.
• 1:39:45
  George: So. So one thing I will say, any time you see a product out of China, it's not crazy low.
• 1:39:51
  Ed: That's my point though.
• 1:39:52
  Propter: Wow.
• 1:39:53
  Ed: That's what. Everyone's pricing it up in that sort of size tier. Like the big thing is that like anthropics, very small model. Haiku is kind of shit. Like if you look at OpenAI, their smaller models are also kind of priced in that same tier. It's just the Chinese models are kicking their ass.
• 1:40:09
  George: Your point is like, for, like that Chinese models are cheaper and better output per dollar in. Right. Substitute substitutable. Fine. Whatever. Your cost of switching to the Chinese model goes down. Right, Right. Okay. This has been true of every industry that China has tried to play in and generally been successful at playing in for the past 20 years. Like that is a subsidized number. That number that Xiaomi is throwing at you is not the real number. That's not the real.
• 1:40:39
  Ed: Of course not. Right, But Open AIs aren't either.
• 1:40:44
  George: No, they're not. But they're closer to the marginal cost than whatever Xiaomi's number is.
• 1:40:48
  Ed: Sure.
• 1:40:49
  George: How about that? OpenAI and Anthropic have received government subsidies in a vague sense of. There is an industrial policy around caring about AI. There's been lots of R and D that's been done on dime and without that we wouldn't have Anthropic or OpenAI. Okay, yes, there have been subsidies to those companies. But compared to the active subsidies that China offers its corporate sector, like, I just. It's not, it's not even in the same realm. So like.
• 1:41:18
  Ed: Right, but how do you square that against like VC? Right? Like they are VCs for the last 20 years have been. I am going to use this word imprecisely. I Apologize if it, if it does not fit. Technically they have been dumping for the last 20 years. Right. Like that's the entire blitzscale model. Like, like I do wonder how much because yeah, like China absolutely subsidizes industries like crazy. But like VCs also subsidized Piperito taxis for a very, very long time.
• 1:41:49
  George: Well, I mean Ed, when did DoorDash IPO.
• 1:41:52
  Ed: DoorDash IPO 2020.
• 1:41:54
  George: Right. Like they were not receiving what I would describe as a quote unquote VC funding anytime for a while up to that. Like, like VC funding is early stage. They're by the time they're getting close to IPO and then IPOing, they are no longer. It's not VC subsidized. The tech industry in general, yes. Has pursued a.
• 1:42:17
  Ed: They were funded at a 400 million Series H in June 2020.
• 1:42:22
  George: Yes, a Series H is no longer VC. You can call it VC if you want.
• 1:42:27
  Ed: I mean fair.
• 1:42:28
  George: It's not VC. Right. And this is like the blurring of public and private that has become so endemic in growth investing. Anthropic is raising at a $900 billion valuation this week per reporting. I don't care who is involved. That's not venture capital anymore.
• 1:42:49
  Ed: Okay, Right.
• 1:42:50
  George: That's front running the ipo. That's all that is.
• 1:42:52
  Ed: That's fair.
• 1:42:53
  George: Like, like that is investing in a well established business with very, very clear benchmarking that is headed for a public market ipo. And you're getting in at a good price because it's good to have friends.
• 1:43:05
  Ed: That makes sense. But Was that true 18 months ago?
• 1:43:11
  George: It's been true for roughly 50.
• 1:43:12
  Ed: No, I mean like was it true for anthropic 18 months ago? Because like how much of this is them? Just like burning cash to build market.
• 1:43:22
  George: To get back to the broader point you're making, which I think is a fair one, that one of the business strategies that tech has pursued, and it's not just VC backed companies, but tech in general has pursued a business strategy of underpricing to stimulate demand. Once you get people locked in and using your service repeatedly, then you raise prices on them. Uber is a great example of this. Doordash all the food delivery companies and that the AI stuff is absolutely as well. The problem is this only works if there is no alternative substitute to. I think Ed has been quite compelling on the idea that there are going to be alternatives. I think it's fair to describe that as a subsidy because you're sort of moving. It's an intertemporal subsidy is sort of the way to think about it.
• 1:44:11
  Ed: Oh, that's an excellent word.
• 1:44:13
  George: Right. You're locking people in with a teaser rate and then anyhow you're banking on
• 1:44:21
  Ed: it being more of a pain in the ass to switch than the money is worth.
• 1:44:24
  George: Right? Exactly, exactly. Whereas the Chinese model is no, just make the marginal cost lower and do that by government subsidy, direct and indirect, mostly indirect, but some direct as well to lower the cost of providing whatever the good or service is. And
• 1:44:43
  Propter: it's some that means working at scale for China is. That's one.
• 1:44:49
  George: That's one aspect of it. Yeah, that's one aspect of it for sure. There are others too. I mean like China can pull a lot of levers to make stuff cheaper. I mean it's just, it's just that simple. Right. There are lots of examples of this, this changing. If you look at some of the online education names, there was an entire economic ecosystem that built up around basically hyper performance for students in China via private companies that got people up to speed better and got you a better score on the test. The Chinese government just decided actually no, we're not doing that anymore. And the industry disappeared overnight. Will they do that with AI? I have no idea. But I would not assume China good quality model for super, super cheap is any more true than OpenAI and Anthropic won't raise prices on you because of the VC subsidy model or that as you put it, VC subsidy demand stimulation model. I think both are uncertain. And this is yet another reason that at the consumer level it may make sense in relatively short order to run everything locally like Proctor was talking about
• 1:45:57
  Ed: because I'm going to be the first one to do it. So I'll tell you how it goes.
• 1:46:00
  George: Yeah, good luck getting the chips or the ram.
• 1:46:03
  Ed: I mean I'm just waiting for, for, for Tim Apple or his designated successor. Goodbye Tim Apple. To give me a box about Gay Big that I can give him $10,000 for.
• 1:46:17
  George: It's not going to be $10,000.
• 1:46:18
  Ed: It's not going to be ten thousand dollars. I just said that because that's what the graphics card costs.
• 1:46:25
  George: The look, the look on low tax's face right now.
• 1:46:28
  Ed: I'm really just shamelessly trying to stretch this out to torment him further.
• 1:46:32
  Lowtax: This has been. I do not know how you three have managed to talk about this nonsense once again for an entire fucking half an hour. This is like every one of our listeners is walking into the ocean right now. If they have not my in laws,
• 1:46:50
  George: buddy, My in laws love this stuff. Okay, but your in law, our biggest
• 1:46:54
  Lowtax: fans, your in laws are obligated to love you, George.
• 1:46:57
  George: No, no, no. They're. They're obligated to tolerate me. They're not obligated to love me. They do love me and I love them too. And it's a wonderful relationship, but, you know, they're not obligated.
• 1:47:06
  Lowtax: Yeah, we're putting you guys on like, I don't know, we'll translate it into tokens so that you three can make sense of it. But we're going to put time caps on this moving forward. Like you guys can have, you guys can have 10 minutes as a little treat, like once every two weeks or something.
• 1:47:25
  Propter: We'll work on a compression and Ed
• 1:47:27
  George: is out of Monster, so I think we're gonna wrap it up right there. It's been an hour and 48 minutes of wall time on this recording. Hopefully the.
• 1:47:35
  Lowtax: I have one more thing. If there are any listeners left who are normal and enjoy funny and horrifying things about our world that normal people enjoy. RFK Jr. RFK Jr. Is walking around Dulles airport holding a starling, which is a type of bird looking for all the world like he's about to do an Ozzy Osbourne to it and just bite the head straight off.
• 1:48:04
  Ed: Where did you find this?
• 1:48:06
  Lowtax: Somebody posted it to Blue Sky. RFK says bring me more HGH Starling rescue today at Dulles Airport. It is a horrifying image. So if you, if you want to hop on to X the Everything app and, and see that you. You may do so.
• 1:48:26
  Propter: Is he just like foraging at airports? Is that.
• 1:48:29
  Lowtax: I. I do not know what he is doing
• 1:48:34
  Ed: the term foraging in an airport. Just. I'm concerned
• 1:48:43
  Propter: Dulles is the least sexy airport in America. I feel strongly about that.
• 1:48:48
  George: Did wait. We're doing sexy airport chat again. How are we doing this two weeks in a row? That's it. We're done.
• 1:48:56
  Propter: We're done.
• 1:48:57
  George: This is the end of the episode. Thank you for listening. Do not rate and review us because after this turd, you're going to give us the worst review possible. We don't want that. We want people to find the podcast. We're not going to rate sexy airports. We're not going to talk about AI anymore for normal men in Charlotte.
• 1:49:12
  Propter: I'm George from D.C. i'm Proctor in outer Florida.
• 1:49:17
  Lowtax: I'm low tech speed run enjoyer in Boston.
• 1:49:20
  George: I'm at Stay normal, y'. All.
• 1:49:22
  Ed: They won't.